Background and Motivation

• Not all rushing yards are equal

• Traditional stats miss the “how” behind yards gained

• Can we “measure” effort using tracking data?

• No universal way to evaluate effort

  • Intangible and subjective
  • Depends on player position, defense faced, game context, snap count/play volume, play call/assignment

Our Data: 2022 NFL Season

• Game, play, player, tracking data from Weeks 1-9 ¹

• Running plays where a running back (RB) is the ball carrier

• Trimmed each play to frames between handoff and end of play

• Backed out key features, aggregated per-play and per-player:

  • Kinetic energy (KE) = 1/2 mv²
  • KE consistency = mean(KE) / sd(KE)
  • Positive work = max {ΔKE, 0}
  • Horizontal displacement of RB in yards

Increasing KE and sustained positive work can lead to high yardage plays

Among our effort metrics, Avg KE best predicts Avg EPA

Early thoughts on methods

• Regression and correlation to determine which metric(s) align best with high impact plays

• Residual effort estimation: predict minimum effort needed to avoid a tackle and compare to actual effort to measure excess (“residual”) effort

• Logistic regression, random forest, etc to predict likelihood of tackle broken or play success from effort and related metrics

• Clustering to classify RBs by effort-reward profile and identify archetypes of RB behavior

Plan of action

• Define research question and scope ✅

• Data cleaning and preprocessing, EDA ✅

• Select effort metric(s)

• Use chosen metric to answer questions related to effort (residual effort, tackle outcome probabilities, minimum required effort per play, etc)

• Interpret results, complete report and presentation